Woods Steven Paul, Childers Meredith, Ellis Ronald J, Guaman Stephanie, Grant Igor, Heaton Robert K
University of California, HIV Neurobehavioral Research Center, Department of Psychiatry (0603-H), 150 West Washington Street, 2nd Floor, San Diego, CA 92103, USA.
Arch Clin Neuropsychol. 2006 Jan;21(1):83-9. doi: 10.1016/j.acn.2005.07.008. Epub 2005 Sep 19.
The accurate interpretation of serial neuropsychological testing is a fundamental activity of neuropsychologists in clinical and research settings. Although prior research supports the construct validity of several individual tests for the reliable measurement of cognitive change, neuropsychologists are often interested in detecting changes in cognitive functioning across a battery of tests. In the present study, we examined the specificity of a modified Reliable Change Index (RCI) methodology applied across a focused battery of commonly used neuropsychological tests. Fifty-seven healthy controls underwent neuropsychological assessment at two time points separated by approximately 1 year. Test-retest reliability coefficients and standard RCI confidence intervals for the individual tests were broadly comparable with prior research in healthy populations. Battery change scores were generated by calculating z-scores of change for each individual test, which were summed across the entire test battery. The RCI methodology was applied to the summed z-score to provide a 90% confidence interval as an indicator of overall cognitive stability. These battery RCI normative standards demonstrated adequate specificity when applied to 29 persons with HIV-1 infection who were classified as medically and neurologically stable. Findings from this study may be useful for both clinicians and researchers seeking normative standards for determining reliable changes in performance across a commonly used battery of neuropsychological tests.
在临床和研究环境中，对系列神经心理学测试进行准确解读是神经心理学家的一项基本工作。尽管先前的研究支持了几种个体测试在可靠测量认知变化方面的结构效度，但神经心理学家通常感兴趣的是在一系列测试中检测认知功能的变化。在本研究中，我们检验了一种改良的可靠变化指数（RCI）方法在一组常用神经心理学测试中的特异性。57名健康对照者在两个时间点接受了神经心理学评估，时间间隔约为1年。各单项测试的重测信度系数和标准RCI置信区间与先前针对健康人群的研究大致相当。通过计算每个单项测试的变化z分数来生成成套测试变化分数，然后将这些分数在整个测试组中相加。将RCI方法应用于相加后的z分数，以提供一个90%的置信区间，作为整体认知稳定性的指标。当将这些成套测试RCI规范标准应用于29名被归类为医学和神经学稳定的HIV-1感染者时，显示出了足够的特异性。这项研究的结果可能对临床医生和研究人员都有用，他们正在寻找规范标准，以确定在一组常用的神经心理学测试中表现的可靠变化。
